A tool to extract specific URLs from different websites

Required a freelance with experience in Linux, Python, HTML parsing, inspecting websites using firebug or similar.


Description of the project:
1. Research, using Firebug or similar tools, how ads look like in a list of websites. They can be from various sources and look different from one another.
2. From the knowledge gained in the research above, create a tool that gets a list of websites and returns a list of all links to those ads, and the size of the image of each ad.


Some important points:
1. The tool you'll provide should run on a Centos7 machine.
2. Suggested language is Python, using libraries that help parse html code.
3. The programmer should understand HTML quite well, in order to understand how to find the right links to the ads.
4. Work involve investigation of different websites and refreshing many times, to learn how to create the tool that can extract the right links. Note that you might need different code for different ads types.
5. The tool should imitate different web browsers on different devices, using different user-agents-strings. We'll supply those strings. To make things simpler. You'll start with the standard Mozilla one, like in the link above.